BlockShuffle¶

class behavenet.data.transforms.BlockShuffle(rng_seed)[source]¶

Bases: behavenet.data.transforms.Transform

Shuffle blocks of contiguous discrete states within each trial.

Methods Summary

__call__(sample)

param sample: dense representation of shape (time)

Methods Documentation

__call__(sample)[source]¶

Parameters: sample (np.ndarray) – dense representation of shape (time)
Returns: output shape is (time)
Return type: np.ndarray